Gap-Filling Model for Electric Grid Wire Path Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electrical power grid modeling approaches fail to accurately predict the locations of important electric grid components such as feeder lines, leading to inaccurate simulations and fault predictions.
Innovation Solution
A system and method utilizing a gap-filling model trained on ground truth images to accurately map paths of electric wires by filling gaps in overhead images, leveraging machine learning techniques such as convolutional neural networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional vector data mapping is used to represent electric grid wires, then the model structure is simple, but the accuracy of predicting wire paths and locations is insufficient
Solution Approach 1:
The patent uses aerial imagery as a visual copy/representation of the actual electric grid wire paths. By training a machine learning model to predict wire paths based on aerial images rather than traditional vector data, the system achieves higher accuracy in locating feeder lines and other grid components. The model learns to identify wire paths by copying the visual patterns from aerial imagery, resolving the contradiction between simple model structure and accurate prediction.
2Measurement precision
If aerial imagery is processed to identify wire paths, then the accuracy of grid component location improves, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary action by pre-processing aerial imagery and training the machine learning model offline before actual grid modeling operations. The model is trained in advance on labeled aerial images to learn wire path patterns, so that during actual operation, the model can quickly predict wire paths without requiring real-time complex processing of aerial imagery. This resolves the contradiction by shifting computational burden to a preliminary training phase.
3Reliability
If machine learning models are trained on aerial imagery to fill gaps in wire paths, then the completeness of grid models improves, but the data processing complexity increases
Solution Approach 1:
The patent introduces an intermediary machine learning model that acts as a bridge between aerial imagery and the final grid model. The model processes aerial images to predict wire paths and fill gaps in the grid model, serving as an intermediary layer that translates visual information into structured grid data. This intermediary approach improves model completeness while managing processing complexity through the use of trained predictive algorithms rather than direct complex image analysis.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a storage device, for filling gaps in electric grid models are enclosed. A method includes obtaining vector data representing first portions of paths of electric grid wires over a geographic region; converting the vector data to first raster image data that depicts an overhead view of the electric grid wires including a first set of line segments representing the first portions of the paths; processing the first raster image data using a gap filling model; obtaining, as output from the gap filling model, second raster image data including a second set of line segments corresponding to gaps included in the input raster image data and representing second portions of paths of the electric grid wires; and converting the second raster image data to vector data representing the first portions and the second portions of paths of the electric grid wires.